Bitcoin Forum

I know this thread probably already exists, but I couldn't find it (sorry)

Anyway, lets assume that BFL ships all it's units, what is a likely difficulty when that happens?

I'm going to say...more?

yes well I figured that, I was looking for an actual estimate, will it double, triple, or will it only increase a little bit.

Oh i understand.. just being funny.  That would be a hard estimate i think.  I'll be interested I what others have to say and their reasoning.

I'd try to figure it out myself, but not exactly sure what goes into calculating difficulty :p

Gonna be looking easily 60 - 100 million.

Well, guess that means I'm out xD
Litecoin here I come :p

After BFL Batch 1 & 2, and Avalon Batch 3, and ASICMiner we'll probably be in the mid 40million range - maybe by June?  Then the oddities like Helveticoin might come in or some other unknowns.  Either way GPUminers will be in Altcoins.  From then on it will be an all out ASIC arms race.  60-100 million by July?  You buy more ASIC if you think BTC will go to 1000USD never knowing what the cap on hashrate will be.  Electric rates will start to matter by this time next year and the Gen2 or even Gen3 units will start coming out.

Or BTC has another issue and something like LTC or PPC takes over - unlikely but you never know.

Sadly Bitcoin doesn't even need an issue of it's own to fail. It's being GOXXED to death. :(

IMO can't really blame Gox, blame all the miners who kept all their bitcoins at Gox and solely used Gox.
But that is a discussion for another thread.

Since no one has any idea how many units are ordered through BFL, no one can say what the difficulty will be when they ship. How many ounces of unicorn blood does it take to fill a faeries bladder? I dunno.

Difficulty scales linearly with Hashrate, if you wish to figure that out, or guess at it. 1200 x 87GH units from Avalon (give or take) will at ~105TH to the existing ~67TH. That will put us roughly into 24Mil difficulty range.

If and when BFL ships it is anybodies guess, but as you can see fairly simply, 150TH will add about 20Mil difficulty, so anywhere between 25mil and 100mil by August seems plausible.

A bit over 300 million.

Tough to say. You can't just multiply the average GH/s of an order by the number of orders and add it to current difficulty. Many people who are spending 100 watts to mine at 500 mh/s will probably drop out, which will reduce the difficulty by some amount as well.

At the moment, it likes the same as before BFL.

Uh, it still is before BFL.  ???

The difficulty is prepare to shell out serious cash to be a Bitcoin miner soon to make anything doing it, and new fierce competition.

My numbers say... 150 mil... but if a new player jumps into the ring it could be double that.

It would be approximately an increase of 0 hashes/second because BFL will never ship.

 ;D

I posted my math on this on my blog. The post is a little long but it shows my method for figuring it out.  Of course (since I ordered a BFL 50ghs) the whole thing assumes that the butterfly lab units will eventually ship. I think they will. If they don't the hash rate won't change much. But if they do ship it changes a LOT.

http://mineshaft.me/2013/04/initial-roi-estimates/

Interesting read. One thing that bothered me while reading through the data presented that perhaps you could clarify. Was network hashrate added by vendors that do currently have a working product(Avalon & ASICMiner) taken into account?

Yes but a guess.  The spreadsheet takes into account the current has rate, which includes some asics already, and was guessing about 20 Ths more by the time BFL ships.  I'll make excel files available soon. I have a better calculation that I plan to post soon.

IC, I missed that tidbit. 20Th may be rather low given that Avalon's batch 2 40Th/s is scheduled to begin shipping sometime between tomorrow and May 5 and ASICMiner is expected to add 5 - 13Th to the network around the same time. I believe what BFL actually has on hand is one wafer of chips only, though perhaps they've made some progress in that area as I do not follow their unofficial releases.

We did it as a shared google doc.  We'll either figure a way to share that with everyone or I'll make an excel doc with it.  But then you can put in any amount for added non-BFL.  And you can put in your order number and see a graph for you.

I was chatting with some other ppl today who said that BFL's first batch was 21TH.  But honestly that would only cover about 700 x 25GH units.  By that measure, they'll be sending one of these "batches" every day or two.

I thought I had heard that BLF-Josh said that anyone who had ordered in the last two weeks would get theirs by July.  But now I can't find that quote again.

I did find this on their website: https://forums.butterflylabs.com/showwiki.php?title=FAQ:Real+Shipping+Estimate (https://forums.butterflylabs.com/showwiki.php?title=FAQ:Real+Shipping+Estimate).


This falls more in line with the guessing that we did.  I'm going to write them and ask some direct questions.  Note that this was a while ago.  They're up to order number 40000 now and I doubt that the 1000 orders a day they're getting now includes ANY cancellations.

Sounds like something he'd say, but I think that assertion is highly suspect given how disorganized they appear to be here in mid April. Josh has also suggested that BFL could assemble and ship hundreds of units per day, but that too seems highly unlikely given their track record and small limited facilities.

I've also noted a conspicuously absent lack of information coming from BFL about their future batches of chips, which is troubling given how quickly they'd deplete the supply they have when shipping commences.

I re-read that faq page.  It wasn't on THEIR faq but the forum faq and that's why it was a guess.  The real faq is where I had seen the thing about July:


And this one which I don't completely understand:


I don't get why they can't "easily" disclose this.  This information is critical to knowing what the mining capability of these boxes are.

Taking them at their word for a minute (which is probably highly optimistic on their part).  Maybe we can reverse engineer how many unit orders they really have.  If they think they're going to produce 400 units a day - that's 2000 a week.  So if they are expecting to be able to ship today's orders by (the end of) July.  Starting 2 weeks from now.  That would be about 8 week to do all the orders.  So 16,000 orders.

That's a lot but it's also not near as many as I had feared when my order number was close to 40000.

I'm not saying they're going to make this schedule.  That's not my point.  But if they were using math to come up with those dates (which people like that do) maybe we can come close to guessing what the numbers in the math were.

Was 400 units/day and 5 day workweeks mentioned by anyone from BFL?

What did they mean by a "unit"?  Jalapeno, Little Single, Single or MiniRig are all considered a "unit"?

I don't know what will be the difficulty. All I know GPU bitcoin mining will be history in this scenario if any miracle not happens to GPU technology and/or energy innovation (solar powered GPUs, etc.)

Yes that number has been mentioned before.  Here is one example: https://forums.butterflylabs.com/bfl-forum-miscellaneous/1736-tour-bfl-offices.html#post23453

Your second question.. that's another key thing.  It's what we were guessing here (full post on my blog like below):

http://mineshaft.me/wp-content/uploads/2013/04/hashrateperbox.jpg

Multiply each guess by the number of units and you have total hash rate.

after bfl runs with the money?

I did a small projection of FPGA farm mining with the rise in difficulty level. Not entirely accurate but it's based off of projections and current USD/BTC, etc.

https://bitcointalk.org/index.php?topic=178051.20 (https://bitcointalk.org/index.php?topic=178051.20)

I did not find one spot where you mentioned anything about difficulty.

Two Jalapenos a day won't have that much influence on difficulty ;D


http://anonymouse.org/cgi-bin/anon-www.cgi/http://img194.imageshack.us/img194/7196/joshlocusts.jpg

Preventing breakdown, perhaps?

Let us assume that hundreds of people have hit them with refunds. And they are now operating in a risky area, where they stay liquid by new preorders and venture money, praying that people won't go crazy asking for refunds, because that would put them in a credit crunch.

The exact credit crunch they need, since they want to pay DHL and UPS.

SO if he does something stupid now, they might go bankrupt, be unable to finish the boxes and several thousand people who are happily mining enough money for a ticket to Kansas will buy one. Happy days ;)

Yeah I think you're right.  Although if they went bankrupt now they wouldn't finish the boxes, and there'd be nothing to go to KS to buy.

I suspect it's not that they can't easily disclose it, but that they don't want to.  I'm just praying that there are a lot of cancellations and not 40,000 units about to be shipped.  Also that it takes them a little while to ship.  That they don't actually produce 400 units a day.  Here's why: the slower the total hashrate rises, the better for those of us who purchased.  The early adopters will get more time to recoup with their boxes.  If they ship in order anyway, the hash rate will be the same when I get my box (barring what is added from other sources) whether they ship slowly or quickly.  But if they ship slowly, and that speed stays slow after my box is shipped, the total hash rate won't rise as fast AFTER I get my box and I too will have more time to recoup.

Not to mention that when these DO ship there will be MORE new orders as people jump on the band wagon thinking that it's a get rich quick scheme.  So I'd rather people stay skeptical for a while.  :)

I think they'd buy a plane ticket to punch them :D



The problem is, they shouldn't even sell that many units. If they pull it off well, there is no TH flood and people will be happy. Happy enough to buy the next 1TH/s miner. Then the 10TH/s miner, once that one is out of order.

There is a limited supply of money coming their way. If they are all smart, they keep making products, but keep the hashrates introduced into the market adding up SLOWLY rather than quickly. That way, the miners will make their money back, everyone is happy.

Market flooding is hard. The 1/3 shipping plan for example is an idiotic idea. Maybe fair to the first buyers, but it will lead to an almost immediate hashrate spike, leading to an immediate hit on the market. This is a singular, disruptive event for the market that is completely idiotic.

If they honour this, take a look at the market prices for days. Volatility is going to spike to some new heights. And then it cools down around a new market price. I am not going to make any predictions where this is though. Only that volatility is gonna skyrocket.

The difficulty rise is in the charts. I estimated a large purchase of FPGAs based off of an old estimation of the rise of difficulty, showing that the rise in difficulty level will make a ~48GH/s almost unprofitable.

It's not that they CAN'T disclose anything about their shipments, it's that they DON'T WANT TO. The lack of transparency that BFL has when it comes to their policies and their business it is downright shameful. I think everyone should come together and threaten a class action lawsuit if they don't release bank statements, produce a bill for their purchase of ASIC chips, give more info shipping, etc. I would bet every dollar I have that they had nothing set up when they started business and then financed development and production with pre-order money. This is illegal, as any company selling something that does not exist must be regulated as a security.

IF YOU SEE THIS, JOSH ZERLAN, YOU'RE A FUCKING SCUMBAG LIAR WHO HAS SUCCESSFULLY CONNED THOUSANDS OF PEOPLE BY LYING TO THEM ABOUT WHEN PRODUCTS WILL SHIP. THE ONLY WAY YOU WILL BE ABLE TO REGAIN FAITH IS BY TELLING YOUR CUSTOMERS THE TRUTH. OPEN UP YOUR RECORDS

God bless this troll in particular. Had a lovely laugh reading that.

Listen, no company owes it's customers such transparency. What could you possibly sue them for "failing to produce a cutting edge product in the time we thought it should be finished" ? Secondly, selling pre-orders on a product that doesn't exist isn't in any way illegal, games aren't "regulated as a security" and people pre-purchase them all the time. In addition to that nobody lied to you, it was well understood that said product didn't exist and would be developed when the pre-orders were sold. In addition to that, you've always had the option of getting a refund if you requested one (although legally you agreed at time of purchase that BFL wasn't obligated to provide a refund). Also, the simple fact that they've refunded folks who've asked should indicate that they didn't infact use the pre-order money for development of the product... and now that they're shipping it's proven to not be 'a scam'.

So while i'm cranky at the delay of the singles and rigs... and that avalon shipped first... I can't blame BFL for being what they are... what we all knew they were from their business practices in the fpga market... a well intending start-up that lacked real world experience in product development, assembly and delivery.

Hopefully they'll get the process ironed out start shipping a couple thousand units a week... then you can find something else to whine about instead of this.

Personally I don't agree. I think BFL should intentionally flood the market with hundreds of units a day for some period of time, at least long enough to catch up on pre-orders. Sure it might crash the exchange rate, but most miners aren't going to be profit taking immediately after they get their rigs up and running... they'll do what they always do, quietly accumulate bitcoins, selling off barely enough to offset power costs and wait for a nice bubble to sell into. After the last speculative bubble I'm not too worried about the price dropping to single digits, there's enough support to keep it 'high enough' and avalon has already increased the hash-rate to the point that we'd probably only see one quick difficulty adjustment instead of the the 3 that were originally projected. and extra 3600 hundred coins for a couple days... not much at all really.

True.

What I meant was, that if they wanted to have this awesome business model, that might be a viable option. Avalon did it right, limiting their orders to a batch of units and then Boom closing the window.

Since they had the floodgates open, I hope they start their "locus" strategy at some time ;)

"Difficulty after BFL"

skyrocket

My guess is 400 mill by years end.

My totally unmotivated guess therefore is 100 times today's difficulty at the end of the year.

Today it is
and thus we will have 1E9 at the end of 2013.

How exactly could it crash the exchange rate?  There are only 3600 coins generated per day (not 3600 hundred which would be 3,600,000.)  25 every 10 minutes.  That's not a constant I know but an average based on the difficulty - which is only adjusted every 2016 blocks.  NOT every 2 weeks.  So if the hash rate were to DOUBLE in one day due to a huge influx of new BFL boxes.. all those blocks would be found much faster.. theoretically twice as fast.. but still there wouldn't be any EXTRA coins out there.  The difficulty would just raise after about a week instead of 2 weeks.

How would that crash the exchange?


If anything.. the exchange rate baseline should raise with the difficulty.  Because new coins generated now require more work and more power.  If the value of coins doesn't raise considerably after the insurgence of BFL boxes, it would no longer be profitable to mine.  When miners can't break even because the difficulty is so high that a $2500 50GHs miner doesn't produce enough coinage - they'll stop mining.  Why power a rig that isn't producing enough coins to pay the electric bill.  Why purchase a $2500 miner if there's no way to recoup the cost of the miner?

So therefore the coins that those miners do produce they will not sell for less than enough to pay for their troubles.  This changes the supply and demand ratio.  As new people become interested in BTC, want to use it for commerce, want to buy it and invest in it, the more BTC we need in circulation to support that demand.  So that would theoretically drive up the price - or cause the need to move a decimal point.  Maybe we no longer care how much 1 BTC trades for but what .01 BTC trades for.  But the raise in demand for new BTC in circulation is supposed to be supplied by miners.  So the value has to go up with the difficulty.  Not necessarily a 1:1 relationship, but there is a practical relationship between the two.

Eat humble waffle!!!

Altcoins are already crashing exchange rate ie through market share.

definitely will go up very quickly, but not immediately, but soon you will be disseminated on a large scale, from 3 months to think about ...

3/4 months... or 5...